Text-based online conferences using Slack
My aim here is to briefly present a way of organizing online academic conferences. I take it that, in 2020, I don’t need to say why this is so, and so I won’t, and will jump straight into logistics, presenting a way to organize asychronous, text-based conferences, using freely available and intuitive software.
I assume a principle of methodological conservatism: that, for the most part, we try to conserve features of the current conference system (gradually increasing, stuffy-room-induced hypoxia and over-caffeination aside) and translate them into an online format, rather than rethinking what a conference ought to be (I am not against such thinking, it’s just not the approach I’m taking here).
In particular, I want to preserve the following features of current conferences:
- collective inquiry
By ephemerality I mean that there are hard limits on the time the conference itself, as well as the talks and other components of the event, takes. Moving online we could, theoretically, remove those limits: we could just recruit a number of people, have them post their papers to some private server, and have them each comment on the others’ papers at their leisure.
But that would arguably be unfortunate. The fact that a given presentation ceases to exist once the speaker stops presenting is perhaps good — it encourages presenting more out-there or underdeveloped ideas without having to be worried about them being available forever. And it encourages on the audience a certain sort of feedback, one that is quickly produced and quickly delivered. While there is need for more considered and slow feedback, I think there’s a good case to be made that this sort of speed-academia is not entirely without merits.
By sociality, I mean that often the most rewarding moments of a conference happen during coffee breaks and dinners. Merely getting a chance to talk to fellow experts is inherently rewarding and should be maintained and encouraged.
Finally, by collective inquiry I mean that in a presentation, the audience members hear the other members’ questions, and often build on them. The best q&a sessions are genuinely instances of group inquiry, and not merely a questioner asking the presenter in the presence of a handful of bystanders.
Before going on to briefly describe a set-up I think could deliver these goals, let me address one big question: why text- as opposed to video- based conferences?
Text-Based Conferences: A Brief Defense
Attention has recently and understandably turned towards online academic events. Most of the ones I’ve seen so far, however, tend to assume that video conferencing will be a part of them. I think that exploring non-video-based options is well worth it, because video places hard constraints on an event. In particular, it requires (or at least almost requires) absolute real-timeness: that presenter and audience be attending to the same thing at the very same time.
If you really think about it, this is big! Indeed, it’s so big that it seems like the best solution we’ve come up to create simultaneous mutual directed attention with so far involves spending thousands of whatever currency we use, flying people all across the world, and putting them in a room together.
But absolute real-timeness doesn’t seem strictly necessary: there is nothing about the nature of academic inquiry that requires it, although I grant that maybe it’s desirable. So: absolute real-timeness is not necessarily but possibly desirable, and we should develop systems that permit this.
Text does. We can, if we so choose, organize a text-based conference such that all participants have to be logged in at the same time. But, and this is the main thing: it allows every other possibility, too. It allows almost-real-time: say, that audience members ask questions within an hour of a given date. Or within four days, or a day, or whatever. That freedom enables us to think: what is the optimum duration of a given presentation — a given instance of group inquiry — both intellectually and practically? Text offers us so many more logistical possibilities than video that, in absence of a very solid reason to the contrary, it should be our default assumption (How does this square with my methodological conservatism? Poorly!)
Slack: Teams, Threads and Channels
Slack is a feature-rich messaging app for co-working. Its users are groups of people or teams, who can post in any of a number of thematic channels. Other users can reply to such posts, creating threads, and there is easy support for importing documents from google drive and other formats.
Then my basic thought is simple: for a text-based conference we have channels corresponding to each paper, as well as a coffee break channel. The team is all the attendees. Authors present their paper by linking to it, and providing a brief introduction before questioning begins. Here is a quick shot of what this might look like:
On the left hand side we have a list of the channels. As I see it, some only become accessible at given times indicated in a publicly available schedule. Assume that I (mipmckeever) am managing the conference. Then I introduce the speaker (here, Herman Cappelen), and provide a link to his paper. Then Cappelen provides a quick summary of it.
The following screen shows how the q&a would work:
You see a question by user ‘tsundell’ and ‘plunkett’ and the thread that tsundell’s question created, with the response from the author. Each question gets its own thread, headed by the question, and followed by the author’s response. As in a normal conference, audience members are encouraged to follow up to questions and replies, and we would implement a mechanism so that a question had to be ‘finished’ before we move on to the next one: we could require, for example, that everyone in the conference react to original question (say, with a ‘thumbs up’ emoji) before we move on to the next question. In the above example, I assumed that we generate a list of questioners which includes at minimum all the invited speakers, but that’s just one example.
Each channel gets automatically archived after a certain period of time, after which its contents are no longer available.
A Worked Example
Let’s consider an example of a possible schedule. Imagine we have a conference with 8 partipants, 4 of whom are on GMT and 4 on EST. We could then allot to each paper a half day, beginning at 2pm GMT, i.e. 9am EST, and going through to 5pm GMT. Each paper gets released at the start of the day, as does the order in which people are to provide questions. After approximately one hour the first person gets a notification that they are due to comment; once they have done so, and the presenter has required, everyone gets a notification that they have done so (I am not sure, but imagine, that this could be automated. Slack looks quite developer-friendly). Others can then chime into the newly created thread, until it is finished and the next person on the list gets a notification that their question is due.
Half way through, there is a coffee break, in which people decamp to the #coffebreak channel, and are actively encouraged to strike up conversations with each other (we could remove some of the social awkwardness these things sometimes involve by encouraging conventions whereby senior people assume responsibility for initiating conversation with junior people).
There is much more to say. I will continue to develop the idea in anticipation of it being needed sooner or later; in the meantime, feedback is very welcome!
Added 23 June 2020: From summer 2020 I’m going to move my occasional writing from medium to tinyletter. If you want to read more from me in your inbox, please consider signing up: https://tinyletter.com/mittmattmutt. I’ll post relatively infrequently, and hopefully interestingly, on the same sort of themes as the blog, so: popular philosophy/explainers, culture, literature, politics/economics, etc. I might also do things like brief reviews of books I read and so on.